The residue I257 at S4-S5 linker in KCNQ1 determines KCNQ1/KCNE1 channel sensitivity to 1-alkanols

Abstract

Aim: KCNQ1 and KCNE1 form a complex in human ventricular cardiomyocytes, which are important in maintaining a normal heart rhythm. In the present study we investigated the effects of a homologous series of 1-alkanols on KCNQ1/KCNE1 channels expressed in Xenopus oocytes.

Methods: ECG recording was made in rats injected with ethanol-containing solution (0.3 mL, ip). Human KCNQ1 channel and its auxiliary subunit KCNE1 were heterologously coexpressed in Xenopus oocytes, which were superfused with ND96 solution; 1-alkanols (ethanol, 1-butanol and 1-hexanol) were delivered through a gravity-driven perfusion device. The slow-delayed rectifier potassium currents IKs (KCNQ1/KCNE1 currents) were recorded using a two-electrode voltage clamp method. Site-directed mutations (I257A) were made in KCNQ1.

Results: In ECG recordings, a low concentration of ethanol (3%, v/v) slightly increased the heart rate of rats, whereas the higher concentrations of ethanol (10%, 50%, v/v) markedly reduced it. In oocytes coexpressing KCNQ1/KCNE1 channels, ethanol, 1-butanol and 1-hexanol dose-dependently inhibited IKs currents with IC50 values of 80, 11 and 2.7 mmol/L, respectively. Furthermore, the 1-alkanols blocked the KCNQ1 channel in both open and closed states, and a four-state model could adequately explain the effects of 1-alkanols on the closed-state channel block. Moreover, the mutation of I257A at the intracellular loop between S4 and S5 in KCNQ1 greatly decreased the sensitivity to 1-alkanols; and the IC50 values of ethanol, 1-butanol and 1-hexanol were increased to 634, 414 and 7.4 mmol/L, respectively. The mutation also caused the ablation of closed-state channel block.

Conclusion: These findings provide new insight into the intricate mechanisms of the blocking effects of ethanol on the KCNQ1 channel.

Figure 1

Electrocardiogram (ECG) changes in rats with acute ethanol treatments. (A) ECG recordings in adult male Wistar rats. Representative tracings of ECG recorded from individual rat before and after injection with 0.3 mL of normal saline or a solution containing 3%, 10% or 50% ethanol (v/v). Each solution was injected into the abdomen of a different rat fixed by two cortical electrodes. The proper position of ECG leads was monitored by the telemetry signal quality using the recording software, and all of the ECG traces were recorded until the quality of the ECG signal was considered acceptable, usually approximately 40 min after each treatment for an individual rat. The ∼1-s ECG recordings were mainly used for average heart rate measurements, albeit the voltage amplitude values being remarkably reduced, especially in the rats that were injected with high concentrations of ethanol (eg, 50%). (B) Comparison of heart rate responses to different treatments (n=3–6). Mean±SEM. ^bP<0.05 vs control. EtOH, ethanol.

Figure 2

Extending the alkyl chain length augments the inhibition efficiency of 1-alkanols on I_Ks currents as recorded in oocytes. (A) Representative currents were recorded from oocytes expressing wild-type KCNQ1/KCNE1 under two-electrode voltage–clamp by voltage steps from −60 to +60 mV in 10-mV increments lasting for 3 s in the presence and absence of 200 mmol/L ethanol (termed as EtOH; A1), 50 mmol/L 1-butanol (BuOH; A2) and 5 mmol/L 1-hexanol (HexOH; A3) as indicated. Voltage protocols included above current traces. (B) Summary plots of I–V curves before and after exposure to different concentrations of EtOH (B1; n=4), BuOH (B2; n=5) and HexOH (B3; n=5) as indicated. All of the currents at each concentration were normalized to their respective control values.

Figure 3

1-Alkanols inhibit I_Ks currents in both the open and closed state of wild-type KCNQ1/KCNE1 channels. (A) Dose-response curves for the open-channel block were plotted for EtOH (A1; n=4), BuOH (A2; n=5) and HexOH (A3; n=5) at +20 mV and were fitted to the Hill equation. Both the IC₅₀ values and the corresponding Hill coefficient n_H are indicated on each panel. (B) Voltage-dependence of IC₅₀ values for EtOH (n=4), BuOH (n=4) and HexOH (n=3). The solid lines are the lines of best fit of the function k(V)=k(0)×exp(−zFV/RT), where k(0) represents the IC₅₀ for 1-alkanols at 0 mV, and z is the fitted value for fractional charges moved during block reaction. The variables of F, R and T have their standard definitions. For EtOH, k(0)=0.0718 with z=−0.4749; k(0)=0.0338 with z=−0.6200 for BuOH; and k(0)=0.0021 with z=−0.5688 for HexOH. (C) Time courses and fractional inhibitions of KCNQ1/KCNE1 currents evoked at +60 mV in the presence of 200 mmol/L EtOH (n=4), 50 mmol/L BuOH (n=4) and 5 mmol/L HexOH (n=5). (D) Dose-dependence of three 1-alkanols effects on the closed state KCNQ1 channels. The inhibition values or ratios by 1-alkanols at the closed state were derived from the values at the intersection of the prolonged line of I_drug/I_control and the zero time axis (t=0 s), as shown in Figure 3C. The solid line represents a Hill equation fit. (E) Normalized conductance-voltage relationship. The conductances that were obtained from steady state currents at the end of the pulse in the presence of various concentrations of EtOH (E1), BuOH (E2) and HexOH (E3) were simultaneously fit to a four-state block model (Model I) as described in the main text (solid lines). The optimal fit for the open and closed-channel block (OB-CB) yielded K_OB=90±8.2 mmol/L, K_CB=60.8±7.0 mmol/L for EtOH (E1), K_OB=20±5.2 mmol/L, K_CB=17.6±6.1 mmol/L for BuOH (E2) and K_OB=5.0±2.1 mmol/L, K_CB=1.6±1.5 mmol/L for HexOH (E3). The error bars represent the SEM. EtOH, ethanol; BuOH, 1-butanol; HexOH, 1-hexanol.

Figure 4

I257, the residue in the intracellular linker of transmembranes S4 and S5, is critical for the 1-alkanol inhibition of KCNQ1 channels. (A) Effects of ethanol on kqt-3 currents. Representative outward current in oocytes expressed with kqt-3 and KCNE1 were evoked by step depolarizations from −90 mV to +60 mV in 10-mV increments in the presence and absence of 200 mmol/L EtOH. The interpulse interval was 3 s. Voltage protocols are presented above current traces. (B) Summary plots of relative currents over voltage before and after exposure to different concentrations of EtOH (n=4), showing that ethanol (200 mmol/L) has a slight effect on kqt-3 currents. (C) An amino acid alignment of the linker of S4 and S5 between C elegans kqt-3 and human KCNQ1. (D) Representative current traces from oocytes that were injected with KCNQ1 (I257A) and KCNE1 in response to voltage steps from −60 to +60 mV in 10-mV increments for 3 s in the presence and absence of 400 mmol/L EtOH (D1), 200 mmol/L BuOH (D2) and 10 mmol/L HexOH (D3) as indicated. (E) Summary plots of relative current over voltage before and after exposure to different concentrations of EtOH (B1; n=4), BuOH (B2; n=5) and HexOH (B3; n=5) as indicated. All of the currents at each concentration were normalized to their respective control values. The error bars indicate the SEM. EtOH, ethanol; BuOH, 1-butanol; HexOH, 1-hexanol.

Figure 5

Substitution of I257 by alanine abrogates the closed-state block of the KCNQ1 channel by 1-alkanols. (A) Dose-response curves for the open-channel block by 1-alkanols; best fitting by Hill's equation resulted in the following: IC₅₀=634±0.8 mmol/L, n_H=0.5 for EtOH (A1, n=5); IC₅₀=414±7.0 mmol/L, n_H=0.5 for BuOH (A2, n=5); and IC₅₀=7.4±0.8 mmol/L, n_H=0.7 for HexOH (A3, n=6). (B) Summary plots of the logarithm of IC₅₀ versus voltage. Solid lines represent the best fits of the function k(V)=k(0)×exp(−zFV/RT), where k(0) represents the IC₅₀ for 1-alkanols at 0 mV, and z is the fitted value for fractional charges moved during the block. The variables F, R and T have their standard definitions. The fits yield the following model parameters: k(0)=0.2053 with z=−0.4010 for EtOH; k(0)=0.0498 with z=−0.6217 for BuOH; and k(0)=0.0052 with z=−0.6999 for HexOH. (C) Normalized peak current amplitudes are plotted as a function of elapsed time. Currents that were encoded with the mutant I257A and KCNE1 were recorded in the presence of 400 mmol/L EtOH, 200 mmol/L BuOH and 10 mmol/L HexOH, respectively. The membrane potential was depolarized to +60 mV lasting for 3 s. (D) Each average C/G_max vs voltage plot with simultaneous fits to the 3-state model for the open channel block (Model II) of I257A with KCNE1 in the presence of various concentrations of EtOH (D1), BuOH (D2) and HexOH (D3). See text for details. The fits correspond to the following parameter values: K_OB=600±18.5 mmol/L for EtOH, K_OB=400±20.1 mmol/L for BuOH and K_OB=2.9±3.2 for HexOH. The error bars indicate the SEM. EtOH, ethanol; BuOH, 1-butanol; HexOH, 1-hexanol.